Crystal Structure of Antagonist Bound Human Lysophosphatidic Acid Receptor 1

Article

Graphical Abstract Authors Jill E. Chrencik, Christopher B. Roth, Masahiko Terakado, ..., Jerold Chun, Raymond C. Stevens, Michael A. Hanson

Correspondence [email protected]

In Brief Structural analyses of a human lysophosphatidic acid receptor reveal plasticity in ligand recognition and suggest mechanisms for modulation of receptors with distinct functions by common ligand scaffolds.

Highlights Accession Numbers d LPA1 structure determined with a selective antagonist 4Z34 4Z35 d Structure-based design led to antagonists with improved 4Z36 characteristics d The LPA1 binding pocket permits extracellular ligand access

unlike related S1P1R d Structural analysis predicts metabolic products of

cannabinoid ligands bind to LPA1

Chrencik et al., 2015, Cell 161, 1633–1643 June 18, 2015 ª2015 Elsevier Inc. http://dx.doi.org/10.1016/j.cell.2015.06.002 Article

Crystal Structure of Antagonist Bound Human Lysophosphatidic Acid Receptor 1

Jill E. Chrencik,1,10 Christopher B. Roth,1 Masahiko Terakado,2 Haruto Kurata,2 Rie Omi,2 Yasuyuki Kihara,5 Dora Warshaviak,3 Shinji Nakade,6 Guillermo Asmar-Rovira,1 Mauro Mileni,1,9 Hirotaka Mizuno,5,6 Mark T. Grifﬁth,1 Caroline Rodgers,1 Gye Won Han,4 Jeffrey Velasquez,4 Jerold Chun,5,8 Raymond C. Stevens,4,7,8 and Michael A. Hanson1,8,11,* 1Department of Structural Discovery, Receptos, San Diego, CA 92121, USA 2Medicinal Chemistry Research Laboratories, Ono Pharmaceutical Co., Ltd., Osaka 618-8585, Japan 3Schro¨ dinger Inc., 120 West 45th Street, New York, NY 10036, USA 4Departments of Biological Sciences and Chemistry, Bridge Institute, University of Southern California, Los Angeles, Los Angeles, CA 90089, USA 5Department of Molecular and Cellular Neuroscience, Dorris Neuroscience Center, The Scripps Research Institute, La Jolla, CA 92037, USA 6Exploratory Research Laboratories, Ono Pharmaceutical Co., Ltd., Ibaraki 300-4247, Japan 7iHuman Institute, ShanghaiTech University, 2F Building 6, 99 Haike Road, Pudong New District, Shanghai, 201210, China 8Co-senior author 9Present Address: Abilita Bio, Inc., San Diego, CA 92121 10Present Address: Structural Biology and Biophysics, Pﬁzer Inc, Groton, CT 06340 11Present Address: GPCR Consortium, San Marcos, CA 92078 *Correspondence: [email protected] http://dx.doi.org/10.1016/j.cell.2015.06.002

SUMMARY with varied fatty acid chain length and saturation. LPA is present in nearly all cells, tissues, and fluids of the body (Mirendil et al., Lipid biology continues to emerge as an area of sig- 2013) and its effects are mediated by cognate cell-surface G pro- nificant therapeutic interest, particularly as the result tein-coupled receptors (GPCRs) consisting of six family mem- of an enhanced understanding of the wealth of bers designated LPA1-LPA6 (Kihara et al., 2014) that couple to signaling molecules with diverse physiological heterotrimeric G protein complexes to activate downstream properties. This growth in knowledge is epitomized signaling pathways. Targeted deletion of the LPA receptors by lysophosphatidic acid (LPA), which functions has revealed physiological effects on every organ system exam- ined thus far, with receptor dysregulation linked to a range of dis- through interactions with at least six cognate G pro- ease indications including hydrocephalus (Yung et al., 2011), tein-coupled receptors. Herein, we present three infertility (Ye et al., 2005), fibrosis (Tager et al., 2008), pain (Inoue crystal structures of LPA1 in complex with antagonist et al., 2004), and cancer (Mills and Moolenaar, 2003). Three of the tool compounds selected and designed through six LPA receptors, including LPA1 (Kihara et al., 2014), are part of structural and stability analyses. Structural analysis a larger lysophospholipid receptor family (the EDG family) that in- combined with molecular dynamics identified a basis cludes the sphingosine 1-phosphate (S1P) receptors, of which for ligand access to the LPA1 binding pocket from the the structure of one member (S1P1) has been reported (Hanson extracellular space contrasting with the proposed et al., 2012). The closely related cannabinoid receptor CB1 access for the sphingosine 1-phosphate receptor. (Hecht et al., 1996; Matsuda et al., 1990) interacts with endogenous ligands structurally related to LPA raising the question of Characteristics of the LPA1 binding pocket raise the possibility of promiscuous ligand recognition of potential polypharmacology between the two receptors, effec- tively connecting these signaling systems. Additional structural phosphorylated endocannabinoids. Cell-based as- details for this family will improve our understanding of their says confirmed this hypothesis, linking the distinct associated function and physiology. receptor systems through metabolically related ligands with potential functional and therapeutic im- RESULTS plications for treatment of disease.

Design of LPA1 Ligands for Structural Studies INTRODUCTION A key for success in GPCR structure determination is the identi- ﬁcation of a ligand that stabilizes the receptor into a single and Lysophosphatidic acid (LPA) is a pleiotropic bioactive lipid pro- stable conformation. To facilitate ﬁnding such a ligand, a stability duced from extracellular lysophospholipids by autotaxin (ATX), assay based on analytical size exclusion chromatography as well as derived from membrane glycerophospholipids by (aSEC) was developed and used to screen a library of com- phospholipases (Aoki, 2004; Hishikawa et al., 2014; Perrakis pounds (Figure S1). The compound used for initial studies of and Moolenaar, 2014), to produce a range of chemical species LPA1, ONO-9780307 (Table S1, Figure S5A and Supplemental

Cell 161, 1633–1643, June 18, 2015 ª2015 Elsevier Inc. 1633 Figure 1. The Overall Structure of LPA1 Colored by Region and Compound Struc- ture in the Bioactive Conformation (A) The ligand ONO-9780307, shown with green carbons, demarcates the binding pocket, which is partially occluded by an N-terminal alpha helix (red) packing closely against the extracellular loops, ECL1 and ECL2 (orange). There are three native disulﬁde bonds in the extracellular region, one of which constrains the N-terminal capping helix to ECL2. (B) The bioactive conformation of ONO-9780307 in green carbons with high torsional strain energy required for positioning the indane ring adjacent to the dimethoxy phenyl ring. Release of this strain with an optimal torsion angle would result in a clash with the binding pocket.

Experimental Procedures), was selected from this library based response to LPA concentration in a calcium mobilization assay on the initial SEC stability assay data (Figure S1)(Tanaka et al., (Figure 2B). To reduce conformational heterogeneity, bRIL was 2011). This assay was used in place of the more traditional modiﬁed by replacing a disordered loop with a short linker

CPM assay (Alexandrov et al., 2008) to allow early access to sta- (mbRIL). Crystals of dsLPA1-mbRIL in complex with ONO- bility data without the need for pure receptor protein. Using in- 3080573 were grown in a LCP cholesterol mixture and diffracted sights gained from the initial structure, several analogs of ONO-9780307 were designed by considering lipophilicity, structural diversity, and potential metabolic stability. These analogs were synthesized (Supplemental Experimental Procedures), tested for stability induction, and used to generate additional co-crystal structures (Table S1). The design feature of ONO- 9910539 (Figure S5B) is an acetyl group for enhanced polar interactions and reduced lipophilicity. The design of ONO-3080573 (Figure S3B) replaced a methylene and the pyrrole ring with an ether and phenyl ring, respectively, to both reduce torsional strain and increase structural diversity and potential metabolic stability.

Structure Determination of LPA1-ONO-9780307, LPA1-ONO-9910539, and LPA1-ONO-3080573 To facilitate crystallization, a thermostabilized b562RIL (bRIL) (Chun et al., 2012), was inserted into the third intracellular loop (3IL) at Ballesteros index positions 5.66 (R233) and 6.24 (R247) (Ballesteros and Weinstein, 1995), and thirty-eight residues were truncated from the carboxyl terminus (C terminus). With this engineered construct (LPA1-bRIL) in complex with compounds ONO-9780307 and ONO-9910539, crystals were produced in LCP supplemented with cholesterol that diffracted to Figure 2. Generation of the Disulfide Engineered LPA1 (dsLPA1- a final resolution of 3.0 A˚ and 2.9 A˚ , respectively, and supported mbRIL) Crystallization Construct (A) Prediction of cysteine mutants replacing Asp2045.37 and Val2826.59 at the structure solution (Figure 1A). Despite similar stability induction tip of TMV and VI, respectively. for ONO-3080573 (Figure S1), we were unable to crystallize (B) Functional analysis of dsLPA1 and full-length wild-type LPA1 through a this compound with the original LPA1-bRIL construct. In order calcium mobilization assay measured in response to LPA. The EC50 value of to further improve stability and reduce conformational heteroge- the stabilized receptor (994 nM) is about 6-fold higher compared to wild-type neity for crystallization in the presence of ONO-3080573, a series (153 nM). Data are presented as a mean (±SEM, n = 3). of disulfide bonds were engineered into various sites within the (C) Thermal stability analysis of dsLPA1-bRIL compared to the non-engineered crystallization construct LPA1-bRIL. The engineered disulfide bond increases extracellular half of the transmembrane region of LPA1-bRIL, the apparent Tm by about 5 C in the presence of ONO-9780307, which was the and one was selected based on superior expression and stability ligand used for stability characterization of new constructs. analysis (dsLPA1-bRIL) (Figures 2A and C). The effect of disulfide (D) Crystal images for dsLPA1-mbRIL in the presence of ONO-3080573. See bond engineering at this site was evaluated using signaling in also Figure S1.

1634 Cell 161, 1633–1643, June 18, 2015 ª2015 Elsevier Inc. Figure 3. Comparison of S1P1 and LPA1 Structural Features

(A). Top-view of LPA1 with structurally divergent

regions of S1P1 (PDB ID: 3V2Y) overlayed (cyan). The configuration of the extracellular region is constrained by two intraloop disulfide bonds in ECL2 and ECL3 and a disulfide bond connecting ECL2 with the N terminus. Although the electron density for the ECL3 loop is poor compared to the rest of the structure, the approximate position could be properly determined and is further from

the N-terminal helix compared to S1P1. In S1P1 Lys285 in this region forms a cation dipole interaction with the C-terminal end of the capping helix,

an interaction that is missing in LPA1. TMI is shifted ˚ 3Acloser to TMVII at the tip compared to S1P1

and ECL0 is a loop in LPA1. Neither ECL0 nor ECL3 are involved in crystal contacts.

(B). The shape of the LPA1 binding pocket is spherical in nature (solid surface) compared to the more linear binding pocket of S1P1 (blue mesh surface). This difference is caused by three changes in the amino acid composition between the two receptors. The

ﬁrst at position 3.33 is an aspartate in LPA1 and a phenylalanine in S1P1. The second at position 5.43 is a tryptophan in LPA1 and a cysteine in S1P1. Finally at position 6.51 LPA1 has a glycine, whereas S1P1 has a leucine. The reduction in side chain bulk at this position opens a sub-pocket occupied by the indane ring of the antagonist series. to 2.9A˚ (Figure 2D). While there are clear differences in experi- of TMI, which is positioned 3 A˚ closer to TMVII compared to mental electron density for the ligands (Figure S2), aside from S1P1 (Figure 3A). This helical repositioning closes a gap between small conformational changes in and around the binding pocket, TMI and TMVII, which was postulated to enable access of hydro- the receptor structures are highly analogous. phobic ligands to the occluded extracellular binding pocket via

the membrane for S1P1. The short segment between TMI and Overall Structural Features of LPA1 in Complex with the N-terminal capping helix (ECL0) is helical in S1P1 but lacks ONO-9780307 secondary structure in LPA1. This change may open access to The initial co-crystal structure of LPA1-bRIL in the antagonist the ligand binding pocket from the extracellular space through state with the compound ONO-9780307 is used for a general increased ﬂexibility and may also contribute to the inward shift description of the receptor. The main fold of the antagonist of TMI through a release of secondary structure steric con- bound LPA1 receptor is similar to other inactive class A GPCRs, straints. Interestingly, ECL1 and ECL2 are comparable between with a canonical seven-transmembrane a-helical bundle (Fig- the two receptors, whereas the position of the third extracellular ˚ ure 1A). The N terminus of LPA1 forms a six-turn alpha helix loop (ECL3) diverges from the S1P1 receptor by up to 8 A at the similar in length and orientation to the equivalent residues in furthest point, resulting in a loss of interactions between this loop

S1P1, but with an additional disulfide bond linking the N-terminal and the rest of the extracellular region (Figure 3A). The overall capping helix to extracellular loop 2 (ECL2), resulting in addi- result is the opening of a ligand access port into the extracellular tional fold stability. The N terminus packs tightly against ECL1 milieu for LPA1, while access from the plasma membrane is and ECL2 and provides charged and polar amino acid side closed (Figure 3A). chains for interactions within the ligand-binding pocket. The These global changes combined with residue substitutions ligand, ONO-9780307, positions its branching aromatic indane generate a binding pocket for LPA1 that is more spherical and dimethoxy phenyl rings adjacent to each other in the spher- compared to S1P1 (Figure 3B). This characteristic supports the ical binding pocket. This positioning requires a strained eclipsed finding that LPA1 has the ability to recognize a more diverse conformation of the torsion angle on the bond adjacent to the repertoire of chemical species of endogenous ligands with acyl indane ring, with a difference of 9.6 kcal/mol between the chains of different lengths and globular conformations beyond observed conformation and the local energy minimum estimated the commonly used LPA 18:1 sn1. Three sequence substitutions by gas-phase quantum mechanics calculations. The ligand can be identified that alter both the shape and polarity of the LPA conformation associated with the local energy minimum is not receptor binding pocket compared to S1P receptors. In LPA re- able to bind in the pocket due to projected clashes with TMVI ceptors, position 3.33 is occupied by an aspartate residue 3.33 (Figure 1B). (Asp129 in LPA1), whereas S1P receptors have a conserved phenylalanine that serves an important role in ligand recognition

Structural Comparison of the LPA1 and S1P1 Receptors (van Loenen et al., 2011). At position 5.43, LPA receptors 5.43 The relative orientation of LPA1 helices is similar to the S1P1 possess a tryptophan residue (Trp210 in LPA1), contrasting receptor, which was expected given the similarity in both with a cysteine residue at this position in most S1P receptors. 6.51 primary sequence (41% identical in the transmembrane region) Finally, LPA receptors have a glycine (Gly274 in LPA1) at po- and recognition of lysophospholipid endogenous ligands (Chun sition 6.51, contrasting with a leucine or alanine that forms impor- et al., 2013). The primary point of divergence occurs at the tip tant van der Waals interactions with the S1P ligands (Figure 3B).

Cell 161, 1633–1643, June 18, 2015 ª2015 Elsevier Inc. 1635 Figure 4. Analysis of the Potential Ligand Entry for S1P1 and LPA1

(A) The position of TMI in S1P1 is shifted away from TMVII with a distance of 11 A˚ as measured between the Ca of Thr481.35 and Glu2947.36, this helical position allows access of ligand (ML056 shown in green carbon sticks) in the

plasma membrane to the binding pocket of S1P1.

(B) The position of TMI in LPA1 is closer to TMVII closing access of ligand (ONO-9780307 shown in green carbon sticks) in the plasma membrane to the

binding pocket. The closer packing of TMI relative to S1P1 may be driven by the presence of Glu3017.43, which forms a hydrogen bond interaction with the backbone of Gly561.39 and Val591.42. (C) Molecular dynamics analysis of the distance between TMI and TMVII measured at four locations represented by a box plot for each measurement with the minimum and maximum indicated by vertical lines. The measure- ments correspond to the distances shown in A˚ monitored over the course of the simulation. The distance between the two helices is consistently larger and

more variable for S1P1 than LPA1 both with and without ML056 and ONO- 9780307 ligands, respectively.

Molecular Dynamics Analysis Supports Ligand Access from Extracellular Space

The divergent position of TMI in LPA1 compared to S1P1 is attributable to both enhanced polar interactions and a substantial reduction in volume at the interface of TMI and TMVII (Figures 4A and 4B). In order to further understand the mode of ligand ac-

cess for both LPA1 and S1P1, comparative molecular dynamics simulations of both receptors were completed, with and without their respective ligands, and the distance between the TMI and TMVII helices was analyzed over the course of three independent 100 ns simulations for each receptor system. Based on this anal-

ysis, it is clear that the distance between TMI and TMVII in LPA1 is indeed smaller and less variable compared to that observed

for S1P1. The distances diverge most noticeably at the extracellular tip of the helices, with a difference of approximately 7 A˚

between LPA1 and S1P1 in the absence of ligand (Figure 4C). Overall, these results strengthen the conclusion that sequence

differences between TMI and TMVII on LPA1 and S1P1 result in an altered ligand access path, whereby LPA1 preferentially re- ceives ligands from the extracellular milieu compared to mem-

brane access for S1P1 ligands.

LPA1 Ligand Binding Interactions with ONO-9780307, ONO-9910539, and ONO-3080573 Despite similar endogenous ligands, similar overall structural characteristics, and mutagenesis studies suggesting simple

interchangeability between LPA1 and S1P1 (Wang et al., 2001), structural comparisons of the pocket shape and polarity revealed substantial divergence between the two receptors. At the top of polar region of the binding pocket, the carboxylic acid of ONO-9780307 interacts through polar and ionic bonds with residues His40, Lys39, and Tyr34, all of which are located on the N-terminal capping helix and ECL0 loop. Across the LPA family, both Lys39 and Tyr34 are highly conserved, how-

ever, His40 is unique to LPA1 and its protonation may be important for high-afﬁnity interactions with the carboxylic acid of ONO-9780307. Although direct experimental support is needed to draw deﬁnitive conclusions, calculations of binding energy

between wild-type LPA1 with a protonated His40 and ONO- 9780307 compared to a His40Ala in silico mutation showed a greater than 1 kcal/mol difference in binding afﬁnity supporting the hypothesis that this position could be partially responsible

1636 Cell 161, 1633–1643, June 18, 2015 ª2015 Elsevier Inc. pocket surface formed by TMIII and TMV presents a signiﬁcantly

enhanced polar environment when compared to S1P1, consisting of an Asp1293.33 on TMIII and Trp2105.43 on TMV, which presents its indole nitrogen for hydrogen bonding interactions with the antagonist ligands (Figure 5A). Ligand design strategies for engaging this polarity while reducing lipophilicity were implemented with ONO-9910539, which features an acetyl group in the para position of the phenyl ring, placing a hydrogen bond acceptor for interaction with the indole nitrogen of Trp2105.43 (Figure 5B). The torsionally strained bioactive conformation of ONO-9780307 helps to ﬁll the large spherical binding pocket, placing the indane ring adjacent to Gly2746.51 on TMVI (Figure 1B and 5A). Ligand design strategies were implemented for ONO- 3080573, which alleviates this torsional strain by replacing the methylene linker with an ether moiety (Figure 5B). This replacement was coupled with replacement of the metabolically labile pyrrole ring found in ONO-9780307 and ONO-9910439 with a phenyl ring, which differentially positions the carboxylic acid, breaking interactions with Arg1243.28, His40, and Tyr34 but gain- ing an interaction with Lys2947.36.

Intersecting Endocannabinoid Pharmacology The additional polarity provided by Asp1293.33 and Trp2105.43 in the hydrophobic binding pocket may have implications for spec- ifying the preference of LPA receptors for long unsaturated acyl chains, serve as a trigger for agonist induced conformational changes, and is interesting in the context of GPCR phylogenetic evolution. The presence of a tryptophan in this position only occurs in 1% of all class A receptors and is unique to the lysophospholipid and cannabinoid receptors (Van Durme et al., 2006). The most abundant endogenous agonist for the

cannabinoid receptor CB1 is 2-arachidonyl glycerol (2-AG; 1,3- dihydroxy-2-propanyl (5Z,8Z,11Z,14Z)-5,8,11,14-eicosatetrae- noate) (Stella et al., 1997; Sugiura et al., 1997). Despite binding Figure 5. Analysis of the Ligand Binding Pocket of LPA1 at a much lower afﬁnity than the other identiﬁed endocannabi- (A) Detailed binding interactions for ONO-9780307 in the LPA1 pocket. Polar noid, anandamide, its higher relative abundance makes 2-AG a interactions are represented by dashed yellow lines, and calculated positions of polar hydrogens are shown to more accurately represent the hydrogen major signaling molecule for cannabinoid receptors (Stella bonding network. The indane binding pocket is formed by Gly2746.51 and et al., 1997). Interestingly, 2-AG is phosphorylated and con- Leu2756.52 which are represented by van der Waals spheres. Areas for verted to 2-arachidonyl phosphatidic acid (2-ALPA) through improving ligand binding are indicated in red text, and red spheres indicate the action of monoacylglycerol (MAG) kinase/acyl glycerol ki- positions for additional polar interactions. nase (AGK) (Kanoh et al., 1986), producing a potential signaling (B) Binding pocket interactions for additional co-crystal structures of ONO- molecule accommodated by the spherical LPA1 binding pocket, 9910539 and ONO-3080573. ONO-9910539 has improved interactions due to although the relative abundance of 2-ALPA has not been inves- the introduction of an acetyl group in the para position of the phenyl ring, which interacts with Trp2105.43. ONO-3080573 reduces the torsional strain induced tigated. Furthermore, an alternative biosynthetic pathway has by the indane position in ONO-9780307 through the replacement of a meth- been reported for the production of anandamide that progresses ylene linker with an ether linkage. In addition replacement of the pyrrole ring through a lysophospholipid analog (pAEA), adding support for with a para substituted phenyl ring alters the interactions of the acid group the idea of metabolic cross-talk between LPA and cannabinoid 7.36 away from the phosphate binding pocket toward Lys294 . See also Fig- receptor systems (Figure 6A) (Liu et al., 2006). ure S2, Tables S1 and S2. The ability of lysophosphatidic acid derivatives of cannabinoid

ligands to bind and activate LPA1 was tested both with 2-ALPA for the observed selectivity of this class of antagonists for LPA1 and pAEA using molecular modeling (Figures 6B and C) and in over LPA2 and LPA3 (Table S2, Figure S3, Figure S4 and Supple- receptor signaling assays previously used to deﬁne LPA1 as an mental Experimental Procedures)(Beard et al., 2013). The polar LPA receptor (Figure 6D). A model for lipid agonist binding binding pocket continues on TMIII, where Arg1243.28 and generated through molecular modeling was used to dock the 3.29 Gln125 form ionic and polar interactions with the carboxylic endogenous ligands into the LPA1 binding pocket. The slight acid and chiral hydroxyl group of ONO-9780307, respectively. expansion of the binding pocket through rotameric shifts of From TMVII, Glu2937.35 stabilizes the position of Lys39 but Trp2105.43 and Trp2716.48, and the exposure of the p clouds of does not directly interact with the ligand (Figure 5A). The binding their respective indole rings enabled favorable interactions with

Cell 161, 1633–1643, June 18, 2015 ª2015 Elsevier Inc. 1637 Figure 6. Pharmacologically Pleitropic Signaling Lipids Overlap through Common Acyl Chain Binding Pockets in the LPA and Cannabinoid Receptors and Enzymatic Interconversion of Polar Head Group (A) The LPA, S1P, and cannabinoid receptors are grouped in the same clade based on a sequence alignment of the transmembrane region. LPA and cannabinoid receptors are proposed to overlap in ligand speciﬁcity when binding polyunsaturated acyl chains delimited by different head group requirements. The biosynthetic map details the conversion of standard phospholipids to both LPA and CB receptor agonists through common pathways. Both known agonists for the cannabinoid system (2-AG and AEA) are one enzymatic step away from LPA ligands via phosphorylation of the head group (2-ALPA and pAEA). (B) Molecular modeling of the endogenous agonist lysophosphatidic acid with an 18 carbon acyl chain (LPA). In the presence of agonist, both Trp2716.48 and Trp2105.43 are predicted to change rotameric states to increase the volume of the binding pocket and direct the respective aromatic p clouds toward the bound ligand. The predicted agonist binding pocket is capable of docking 2-ALPA and pAEA, which are generated through enzymatic action from endogenous cannabinoids. (legend continued on next page)

1638 Cell 161, 1633–1643, June 18, 2015 ª2015 Elsevier Inc. the double bonds of the phosphorylated cannabinoid ligands creases ligand binding affinity by more than 1 kcal/mol (Beard (Figure 6B). The ability of both 2-ALPA and pAEA to bind in the et al., 2013). If substantiated through direct experimental obser- modeled agonist pocket without significant ligand strain, sup- vation, this could further support the hypothesis that the ports the notion that the hydrophobic binding pockets of LPA1 neoplastic activity of LPA1 is enhanced (Hama et al., 2004; Haus- and CB1 can indeed bind the same, poly-unsaturated acyl chains mann et al., 2011; Mills and Moolenaar, 2003; Perrakis and Moo- with metabolically interconvertible head groups. lenaar, 2014) in the acidic environment associated with hypoxic The molecular modeling provided sufficient rationale for tumors (Herr et al., 2011; Justus et al., 2013; Kim et al., 2006; testing both 2-ALPA and pAEA in LPA1 signaling assays. Tannock and Rotin, 1989; Wojtkowiak et al., 2011). Structural an- Remarkably, all phosphorylated ligands produced robust alyses predicts a functional complementarity and/or redundancy signaling as judged by calcium response and cAMP inhibition between CB1 and LPA1 depending on the balance of phosphor- in stable cell lines heterologously expressing LPA1 (Fukushima ylation states for AEA and 2-AG in particular, since it is a favored et al., 1998). Further, a neurite retraction assay, which measures substrate of MAG kinase, at least in the brain (Kanoh et al., 1986). cell morphology changes mediated by LPA1, showed robust The interplay of these receptor-ligand systems may contribute to activation by pAEA (Figure 6D). Neurite retraction involves acti- phenotypic variability seen among LPA1 knockout strains and vation of the small GTPase Rho that can also activate F-actin variable null mutant survival within susceptible strains that are stress fibers in adherent cells, and this classic response was attributable to brain phenotypes (Contos et al., 2000; Matas- also activated by pAEA. Together these results highlight the Rico et al., 2008), since brain expression of both receptors, along selectivity of the phosphate headgroup on LPA1 in addition to with ligand metabolism, is widespread and can overlap as the promiscuity of the binding site for acyl chain geometry and occurs in oligodendrocytes (Benito et al., 2007; Weiner et al., degree of saturation, wherein chemically distinct ligands that 1998). Complementary access to LPA1 through phosphorylated are capable of accessing this space can produce common re- CB1 ligands—and vice versa—represent examples of a single sponses of receptor activation and down-stream signaling. molecular species that could serve as both a primary receptor modulator and simultaneous prodrug for a different receptor, DISCUSSION as represented by the endocannabinoids acting at their receptors, with phosphorylation producing selective LPA The growing family of GPCRs mediating membrane-derived receptor ligands. Currently, compounds developed to target lysophospholipid signals (Kihara et al., 2014) is remarkable for LPA receptors have entered Phase II clinical trials for pulmonary its ability to respond to diverse changes in the cellular milieu fibrosis and systemic sclerosis. Compounds exhibiting antago- through multiple signaling mechanisms, and to discriminate nist and/or agonist activity for both the CB1 and LPA1 receptors structurally similar molecules. The ability of antagonist structures have yet to be tested but may be superior for certain indications. to inform receptor agonist binding is atypical, yet may reflect the evolutionary selection of binding pockets for lipid ligands that are EXPERIMENTAL PROCEDURES produced from precursor reservoirs within which the receptors are embedded, which may further contribute to constitutive or in- Ligand Selection Optimization and Stability Analysis verse agonist activities (Bond and Ijzerman, 2006). A somewhat Compounds based on a 3-pyrrolylpropionic acid core were selected from a surprising result was the structural preference of LPA1 for ligand library of previously developed LPA1 antagonists based on superior receptor access from the extracellular milieu, rather than via the mem- stability induction properties (Tanaka et al., 2011). From this initial screen, ONO-9780307 was selected for structural determination. Analysis of the bind- brane as was reported for S1P1. Nevertheless, this structural ing pocket revealed three regions for improved interactions. ONO-9910539 prediction is supported by the biological existence and routine was designed to improve polar interactions with Trp2105.43 through the intro- experimental use of albumin bound LPA (Postma et al., 1996) duction of an acetyl moiety on the phenyl ring, and ONO-3080573 was de- to activate LPA1 (Fukushima et al., 1998), and is further consis- signed to reduce torsional strain associated with the bioactive conformation tent with the proposed delivery of LPA by its major extracellular of the scaffold through the introduction of an ether linkage to the indane ring biosynthetic enzyme, autotaxin, via a hydrophobic channel iden- system (Figure 5B, see also Supplemental Experimental Procedures, Figures tified in its crystal structure (Nishimasu et al., 2011). The structure S3B and S5 and Table S1). To measure the effect of compounds on the stability of the receptor, each of LPA identified binding interactions between antagonist and 1 compound was tested using a fluorescence based SEC stability assay. Briefly, His40, having a calculated pKa of 6.8 and 4.5 in the presence after heating the GFP-tagged protein in the presence of ligand, an SEC profile and absence of ligand, respectively. Although speculative, of each data point was analyzed for fraction of folded protein, which was computational analysis indicates that full protonation of His40 in- plotted as a function of temperature or time to determine approximate melting

(C) Conformational root mean square ﬂuctuation of LPA over the course of a 100 ns molecular dynamics simulation. The orientation of the ligand is consistent throughout the calculation indicating a reasonable starting point for modeling agonist interactions.

(D) Analysis of intracellular signaling and cellular functions in B103 cells heterologously expressing full-length wild-type LPA1. The data are presented as mean normalized values (±SEM) to maximum responses induced by 3 mM 1-oleoyl-LPA (n = 3) for Ca2+ mobilization assay, and by 10 mM forskolin (n = 3) for cAMP 2+ assay. For the Ca mobilization assay EC50 values for LPA 18:1 sn1 (n = 3), 1-ALPA (n = 3), 2-ALPA (n = 2), and pAEA (n = 2) are 146 ± 74 nM, 1,125 ± 1,484 nM,

98 ± 143 nM, and 2,120 ± 3,885 nM, respectively. For the cAMP assay EC50 values for LPA 18:1 sn1 (n = 2), 2-ALPA (n = 2), and pAEA (n = 2) are 0.345 ± 0.202 nM, 0.302 ± 0.54 nM, and 1.50 ± 2.04 nM, respectively. Neurite retraction was quantified under microscopic observation. Filamentous (F)-actin stress fibers, produced by Rho GTPase activation like neurite retraction, were also produced by pAEA stimulation. The representative cell morphologies were shown (scale bars, 50 mm). The F-actin stress fiber formation was observed after 15 min treatment with LPA and pAEA (scale bars, 10 mm).

Cell 161, 1633–1643, June 18, 2015 ª2015 Elsevier Inc. 1639 lular loop at positions V232 and R248. Baculovirus directed expression in Table 1. Crystallographic Data Collection and Refinement Spodoptera frugiperda (Sf9) insect cells was driven by the ie1GP64 promoter Statistics with a cleavable HA signal sequence on the N terminus of the receptor and a Data Collection and Refinement Statistics cleavable purification tag consisting of a 103 poly histidine tag upstream of a

Ligand ONO- ONO- ONO- FLAG epitope tag. The resulting construct (LPA1-bRIL) was expressed in 9780307 9910539 3080573 Spodoptera frugiperda (Sf9) insect cells using the Bac-to-Bac baculovirus expression system (Invitrogen) at a yield of 1 mg/l and purified using Resolution Cutoff (A˚ ) 3 2.9 2.9 IMAC chromatography (see also Supplemental Experimental Procedures). Number of Crystals 86 53 39 Protein samples of LPA1-bRIL and dsLPA1-mbRIL were reconstituted into Data Collection Statistics lipidic cubic phase (LCP) and crystallization set-ups were performed at room temperature (20C). Data-collection quality crystals were obtained Space Group P2 2 2 P2 2 2 P2 2 2 1 1 1 1 1 1 1 1 1 over a period of 30 days from precipitant condition containing 0.1 M sodium Cell Dimensions 34.3, 112.2, 34.6, 112.4, 34.4, 111.9, citrate (pH 5.5), 34% (À38% [v/v] PEG400 and 200 mM ammonium acetate; a,b,c (A˚ ) 154.6 155.7 154.0 Figure 2D). Crystals were harvested directly from LCP matrix and flash frozen Number of Reflections 42,304 27,843 69,138 in liquid nitrogen for diffraction analysis. X-ray diffraction data were collected Measured on the 23ID-D/B beamline (GM/CA-CAT) at the Advanced Photon Source (Argonne, IL) using a 10 mm diameter minibeam with a MarMosaic 300 Number of Unique 10,599 10,883 12,701 CCD detector. A rastering and data-collection strategy was followed as pre- Reflections viously described (Cherezov et al., 2009; Hanson et al., 2012). Unattenuated Resolution (A˚ ) 45 - 3.0 47 - 2.9 47 - 2.9 exposures of 5–10 s per degree were required to observe diffraction to 3A˚ (3.16 - 3.0) (3.1 - 2.9) (3.1 - 2.9) resolution. Due to the limited amount of data collected per crystal, a protocol similar to that developed for S1P diffraction analysis was utilized to R 0.19 (0.46) 0.14 (0.49) 0.13 (0.59) 1 merge assemble the data. After data integration and scaling the structures were Mean I/s(I) 4.7 (1.4) 4.0 (1.0) 7.2 (1.3) solved using molecular replacement followed by multiple cycles of refine- Completeness 85 (75) 78 (62) 92 (80) ment and rebuilding in order to generate the final model (see Supplemental Experimental Procedures). Statistics are presented in the crystallographic Redundancy 4.0 (2.1) 2.6 (1.7) 5.4 (2.3) summary table (Table 1). CC1/2 0.99(0.27) 0.99(0.38) 1.0(0.56) Refinement Statistics Stability Engineering Using Disulfide Bond Addition Candidates for disulfide bond engineering were generated computationally Resolution (A˚ ) 30 - 3.0 30 - 2.9 30 - 2.9 using algorithms implemented in BioLuminate (Schro¨ dinger, 2012) and similar Number of Reflections 10,576 (531) 11,627 (584) 12,660 (625) to previously described procedures (see also Supplemental Experimental Pro- (test set) cedures)(Salam et al., 2014). Five potential candidates were identified on the extracellular half of the receptor consisting of two cysteine mutations per site Rwork /Rfree (%) 25.4 / 28.1 26.5 / 27.9 27.4 / 29.1 to facilitate the formation of an additional disulfide bond. After generation of the Number of mutant constructs and testing for expression, the double mutant D204C and Atoms V282C was selected for crystallization studies. This candidate was selected Protein 2,991 2,977 3,014 due to the enhanced expression profile as compared to the other four con- Ligand 38 41 38 structs, which showed little to no expression, as well as an enhanced thermal stability profile as characterized by the SEC thermal stability assay using highly Lipid and other 18 21 21 purified receptor without a GFP tag (see also Supplemental Experimental Average B Factor (A˚ 2) Procedures). Protein 73.2 95.1 87.8 Molecular Modeling and Dynamics Studies bRIL 95.5 114.8 121.7 Crystal structures of the S1P1 (PDB ID: 3V2Y) and LPA1 were prepared using Ligand 48.6 70.9 63.7 the Schro¨ dinger Protein Preparation Wizard (Sastry et al., 2013) and Lipid and other 62.3 82.2 64.1 embedded into a pre-equilibrated POPC (1-palmytoil-2-oleoyl-sn-glycero- 3-phosphatidylcholine) lipid bilayer and solvated with water using the Des- RMSD mond System Builder. Chloride ions were added to neutralize the system Bond Lengths (A˚ ) 0.01 0.009 0.009 and the ionic strength was set to 0.15 M. The OPLS 2.1 force field (Schro¨ - Bond Angles (o) 1.08 0.92 0.87 dinger, 2014d)(Shivakumar et al., 2012) was utilized for all calculations and systems were relaxed using the default membrane relaxation protocol Ramachandran Plot Statistics (%) as implemented within the Schro¨ dinger package. Molecular Dynamics simu- Favored Regions 96.3 96.8 97.3 lations were performed in the NPT ensemble at 300K and 1 atm pressure Allowed Regions 3.7 3.2 2.7 using Desmond (D.E. Shaw Research, 2014; Schro¨ dinger, 2014c). The two receptors were analyzed both with and without ligands. Van der Waals and Disallowed Regions 0 0 0 short range electrostatic interactions were cut off at 9 A˚ and smooth particle Data for high resolution shells is shown in parenthesis where applicable. mesh Ewald (PME) (Essmann et al., 1995) method was employed for calculation of long range electrostatic interactions. The reversible reference system propagation algorithm (RESPA) multiple time step approach was used temperature or half-life, respectively (see also Supplemental Experimental with a time step of 2 ps and long-ranged electrostatic interactions were Procedures and Figure S1). computed every 6 ps (Tuckerman et al., 1992). The temperature and pressure were controlled using a Nose-Hoover chain thermostat (Martyna

Crystallization and Diffraction Analysis of LPA1-bRIL and et al., 1992) and Martyna-Tobias-Klein (Martyna et al., 1994) barostat, dsLPA1-mbRIL respectively. Three independent simulations of 100 ns each were performed

Synthetic cDNA encoding the full-length LPA1 receptor was generated with a for each system on a Nvidia GTX GPU card with coordinates saved every stabilized apocytochrome c (bRIL) protein inserted within the third intracel- 1.2 ps for subsequent analysis. Molecular dynamics used for docking and

1640 Cell 161, 1633–1643, June 18, 2015 ª2015 Elsevier Inc. modeling studies utilized similar system parameters but with a 1.2 ns simu- SUPPLEMENTAL INFORMATION lation. Coordinates were saved every 1.2 ps and clustered by root mean square deviation of the residues within a 4 A˚ radius of the ligand. Represen- Supplemental Information includes Supplemental Experimental Procedures, tative structures of each cluster were tested for ability to dock endogenous five figures, and two tables and can be found with this article online at http:// ligands with minimal torsional strain. dx.doi.org/10.1016/j.cell.2015.06.002. Change in binding affinity toward the ligand ONO-9780307 caused by either deprotonation of the residue His40 or its mutation to alanine was calculated AUTHOR CONTRIBUTIONS using the Residue Scanning functionality in BioLuminate (Schro¨ dinger, 2014a) which incorporates the Prime MM-GBSA approach (Beard et al., J.E.C. led the structure determination efforts. C.B.R. led the construct devel- 2013). A cutoff of 1 kcal/mol was used as a threshold for defining potential opment and expression efforts. D.W. performed the molecular dynamics sim- effect from perturbation. ulations. M.T. and H.K. designed and synthesized antagonists. R.O. assisted Quantum mechanical calculations for determining the energy penalty asso- in structural analysis for compound optimization efforts. S.N. evaluated biolog- ciated with the indane position observed in the crystal structures were carried ical effect of antagonists. G.A.R. performed the ligand stability screening and out after geometry optimizations using Jaguar 8.6 (Schro¨ dinger, 2014b) at the assisted in crystallization efforts. M.M. and M.T.G. assisted in purification and B3LYP/6-31G** level of theory with and without torsional constraints on the in- crystallization efforts. G.W.H. finalized the structural refinement and quality dane ring region of the compound. control of the crystallization data. J.V. performed cloning for the functional analysis studies. C.R. developed large-scale expression protocols. Y.K. and Functional Analysis Studies H.M. carried out the functional analyses, and with J.C., conceived and wrote

HA-tagged full-length human receptor (LPA1) and disulfide stabilized full- portions of the manuscript. R.C.S. conceptualized the project and wrote por- length human receptor (dsLPA1), DNAs were constructed and cloned into tions of the manuscript. M.A.H. led the project and wrote the main manuscript. pCXN2.1 (Niwa et al., 1991). Rat neuroblastoma B103 cells, maintained in complete media were transfected with each plasmid using GeneJET Plus ACKNOWLEDGMENTS (Thermo Scientific) (Fukushima et al., 1998). After 24 hr incubation, media was replaced with complete media containing 1 mg/ml G418 (Life Technolo- The authors thank Angela Walker for critical review of this manuscript and gies) and the cells were cultured for an additional 2 weeks. Cells labeled J. Smith, R. Fischetti, and N. Sanishvili for assistance in development and with anti-HA antibody (Roche) followed by PE-conjugated anti rat IgG use of the minibeam and beamtime at GM/CA-CAT beamline 23-ID at the (eBioscience) as primary and secondary antibodies, respectively, were sorted Advanced Photon Source, which is supported by National Cancer Institute by FACS Aria II (BD Biosciences) using a PE-highly-positive gate, collected, grant Y1-CO-1020 and NIGMS grant Y1-GM-1104. This work was supported and cultured in complete media containing Penicillin-Streptomycin (Life Tech- in part by the NIH grants MH051699, NS082092 and NS084398 to J.C., and nologies) to create stable lines. U54 GM094618 (target GPCR-235) and ShanghaiTech University to R.C.S. For Ca2+ measurement, the stable cell lines were starved with FreeStyle 293 Expression Medium (Life Technologies) for at least 1 hr before loading dye Received: December 3, 2014 according to the manufacture’s instruction with minor modifications (FLIPR Revised: February 17, 2015 Calcium 4 Assay Kit; Molecular Device). Intracellular Ca2+ mobilization was Accepted: May 27, 2015 monitored in response to increasing ligand concentration using a scanning Published: June 18, 2015 fluorimeter (FLEXstation 3; Molecular Devices) at an excitation wavelength of 485 nm and an emission wavelength of 525 nm. For cAMP assay, stable cell lines were starved with FreeStyle 293 Expres- REFERENCES sion Medium (Life Technologies) overnight. The cells were incubated with ligands for 30 min at room temperature in Hank’s Balanced Salt Solution con- Alexandrov, A.I., Mileni, M., Chien, E.Y., Hanson, M.A., and Stevens, R.C. taining 10 mM forskolin and 0.5 mM 3-isobutyl-1-methylxanthine. Intracellular (2008). Microscale fluorescent thermal stability assay for membrane proteins. cAMP levels were determined using a cAMP-Screen System (Applied Bio- Structure 16, 351–359. systems) according to the manufacturer’s instructions. Aoki, J. (2004). Mechanisms of lysophosphatidic acid production. Semin. Cell For neurite retraction and stress fiber formation assay, stable cell lines Dev. Biol. 15, 477–489. were starved with FreeStyle 293 Expression Medium (Life Technologies) Ballesteros, J.A., and Weinstein, H. (1995). Integrated methods for the con- overnight. The cells were incubated with ligands for 15 min at 37 C, washed struction of three dimensional models and computational probing of struc- and then fixed with 2% paraformaldehyde. Cells with round morphology ture-function relations in G-protein coupled receptors. Methods Neurosci. lacking neurite extensions were counted and represented as a percentage 25, 366–428. of the observed cells. The f-actin stress fiber formation was studied using Beard, H., Cholleti, A., Pearlman, D., Sherman, W., and Loving, K.A. (2013). rhodamine phalloidin staining. Fixed cells were washed with 0.3% Triton Applying physics-based scoring to calculate free energies of binding for single X-100 in PBS, and stained with 1 mg/ml rhodamine-phalloidin for 30 min at amino acid mutations in protein-protein complexes. PLoS ONE 8, e82849. room temperature. After washing, the cells were mounted using Vectorshield (Vector Laboratory). Images were captured using a Nikon C2 laser-scanning Benito, C., Romero, J.P., Tolo´ n, R.M., Clemente, D., Docagne, F., Hillard, C.J., confocal microscope. Guaza, C., and Romero, J. (2007). Cannabinoid CB1 and CB2 receptors and Standard lipids were obtained from commercial sources: 1-oleoyl-LPA and fatty acid amide hydrolase are specific markers of plaque cell subtypes in hu- 1-arachidonoyl-LPA (Avanti Polar Lipids), AEA, 2-AG and pAEA (Cayman man multiple sclerosis. J. Neurosci. 27, 2396–2402. Chemical). Enzymatic synthesis was employed for 2-arachidonoyl-LPA that Bond, R.A., and Ijzerman, A.P. (2006). Recent developments in constitutive re- was enzymatically synthesized from 1,2-diarachidonoyl-sn-glycero-3-phos- ceptor activity and inverse agonism, and their potential for GPCR drug discov- phate (diarachidonoyl-PA) (Avanti Polar lipids) using Rhizopus arrhizus lipase ery. Trends Pharmacol. Sci. 27, 92–96. (Sigma) (see also Supplemental Experimental Procedures). Cherezov, V., Hanson, M.A., Griffith, M.T., Hilgart, M.C., Sanishvili, R., Nagar- ajan, V., Stepanov, S., Fischetti, R.F., Kuhn, P., and Stevens, R.C. (2009). Ras- ACCESSION NUMBERS tering strategy for screening and centring of microcrystal samples of human membrane proteins with a sub-10 microm size X-ray synchrotron beam. Coordinates and structures factors have been deposited in the Protein Data J. R. Soc. Interface 6 (5), S587–S597. Bank for LPA1-ONO9780307, LPA1-ONO9910539 and LPA1-ONO3080573 Chun, E., Thompson, A.A., Liu, W., Roth, C.B., Griffith, M.T., Katritch, V., (PDB: 4Z34, 4Z35, 4Z36). Kunken, J., Xu, F., Cherezov, V., Hanson, M.A., and Stevens, R.C. (2012).

Cell 161, 1633–1643, June 18, 2015 ª2015 Elsevier Inc. 1641 Fusion partner toolchest for the stabilization and crystallization of G protein- Matas-Rico, E., Garcıá-Diaz, B., Llebrez-Zayas, P., Lo´ pez-Barroso, D., Santıń, coupled receptors. Structure 20, 967–976. L., Pedraza, C., Smith-Ferna´ ndez, A., Ferna´ ndez-Llebrez, P., Tellez, T., Re- Chun, J., Hla, T., Spiegel, S., and Moolenaar, W.H. (2013). Lysophospholipid dondo, M., et al. (2008). Deletion of lysophosphatidic acid receptor LPA1 re- Receptors: Signaling and Biochemistry (Hoboken, New Jersey: John Wiley & duces neurogenesis in the mouse dentate gyrus. Mol. Cell. Neurosci. 39, Sons, Inc.), pp. i–xviii. 342–355. Contos, J.J., Fukushima, N., Weiner, J.A., Kaushal, D., and Chun, J. (2000). Matsuda, L.A., Lolait, S.J., Brownstein, M.J., Young, A.C., and Bonner, T.I. Requirement for the lpA1 lysophosphatidic acid receptor gene in normal suck- (1990). Structure of a cannabinoid receptor and functional expression of the ling behavior. Proc. Natl. Acad. Sci. USA 97, 13384–13389. cloned cDNA. Nature 346, 561–564. D.E. Shaw Research (2014). Desmond Molecular Dynamics System, version Mills, G.B., and Moolenaar, W.H. (2003). The emerging role of lysophosphati- 3.9 (New York: D.E. Shaw Research). dic acid in cancer. Nat. Rev. Cancer 3, 582–591. Mirendil, H., Lin, M.-E., and Chun, J. (2013). Lysophosphatidic Acid (LPA) Essmann, U., Perera, L., Berkowitz, M.L., Darden, T., Lee, H., and Pedersen, Receptor Signaling. In Lysophospholipid Receptors: Signaling and Biochem- L.G. (1995). A smooth particle mesh Ewald method. J. Chem. Phys. 103, istry (John Wiley & Sons, Inc.), pp. 1–39. 8577–8593. Nishimasu, H., Okudaira, S., Hama, K., Mihara, E., Dohmae, N., Inoue, A., Ish- Fukushima, N., Kimura, Y., and Chun, J. (1998). A single receptor encoded itani, R., Takagi, J., Aoki, J., and Nureki, O. (2011). Crystal structure of auto- by vzg-1/lpA1/edg-2 couples to G proteins and mediates multiple cellular re- taxin and insight into GPCR activation by lipid mediators. Nat. Struct. Mol. sponses to lysophosphatidic acid. Proc. Natl. Acad. Sci. USA 95, 6151– Biol. 18, 205–212. 6156. Niwa, H., Yamamura, K., and Miyazaki, J. (1991). Efficient selection for high- Hama, K., Aoki, J., Fukaya, M., Kishi, Y., Sakai, T., Suzuki, R., Ohta, H., Yamori, expression transfectants with a novel eukaryotic vector. Gene 108, 193–199. T., Watanabe, M., Chun, J., and Arai, H. (2004). Lysophosphatidic acid and autotaxin stimulate cell motility of neoplastic and non-neoplastic cells through Perrakis, A., and Moolenaar, W.H. (2014). Autotaxin: structure-function and LPA1. J. Biol. Chem. 279, 17634–17639. signaling. J. Lipid Res. 55, 1010–1018. Hanson, M.A., Roth, C.B., Jo, E., Griffith, M.T., Scott, F.L., Reinhart, G., De- Postma, F.R., Jalink, K., Hengeveld, T., Bot, A.G., Alblas, J., de Jonge, H.R., sale, H., Clemons, B., Cahalan, S.M., Schuerer, S.C., et al. (2012). Crystal and Moolenaar, W.H. (1996). Serum-induced membrane depolarization in structure of a lipid G protein-coupled receptor. Science 335, 851–855. quiescent fibroblasts: activation of a chloride conductance through the G protein-coupled LPA receptor. EMBO J. 15, 63–72. Hausmann, J., Kamtekar, S., Christodoulou, E., Day, J.E., Wu, T., Fulkerson, Salam, N.K., Adzhigirey, M., Sherman, W., and Pearlman, D.A. (2014). Struc- Z., Albers, H.M., van Meeteren, L.A., Houben, A.J., van Zeijl, L., et al. (2011). ture-based approach to the prediction of disulfide bonds in proteins. Protein Structural basis of substrate discrimination and integrin binding by autotaxin. Eng. Des. Sel. 27, 365–374. Nat. Struct. Mol. Biol. 18, 198–204. Sastry, G.M., Adzhigirey, M., Day, T., Annabhimoju, R., and Sherman, W. Hecht, J.H., Weiner, J.A., Post, S.R., and Chun, J. (1996). Ventricular zone (2013). Protein and ligand preparation: parameters, protocols, and influence gene-1 (vzg-1) encodes a lysophosphatidic acid receptor expressed in on virtual screening enrichments. J. Comput. Aided Mol. Des. 27, 221–234. neurogenic regions of the developing cerebral cortex. J. Cell Biol. 135, 1071–1083. Schro¨ dinger (2012). BioLuminate version 1.0 (New York: Schro¨ dinger, LLC). Herr, K.J., Herr, D.R., Lee, C.W., Noguchi, K., and Chun, J. (2011). Stereotyped Schro¨ dinger (2014a). BioLuminate version 1.7 (New York: Schro¨ dinger, LLC). fetal brain disorganization is induced by hypoxia and requires lysophosphati- Schro¨ dinger (2014b). Jaguar 8.6 (New York: Schro¨ dinger, LLC). dic acid receptor 1 (LPA1) signaling. Proc. Natl. Acad. Sci. USA 108, 15444– Schro¨ dinger (2014c). Maestro-Desmond Interoperability Tools, version 3.9 15449. (New York: Schro¨ dinger, LLC). Hishikawa, D., Hashidate, T., Shimizu, T., and Shindou, H. (2014). Diversity and Schro¨ dinger (2014d). OPLS2.1 (New York: Schro¨ dinger, LLC). function of membrane glycerophospholipids generated by the remodeling Shivakumar, D., Harder, E., Damm, W., Friesner, R.A., and Sherman, W. pathway in mammalian cells. J. Lipid Res. 55, 799–807. (2012). Improving the Prediction of Absolute Solvation Free Energies Using Inoue, M., Rashid, M.H., Fujita, R., Contos, J.J., Chun, J., and Ueda, H. (2004). the Next Generation OPLS Force Field. J. Chem. Theory Comput. 8, 2553– Initiation of neuropathic pain requires lysophosphatidic acid receptor 2558. signaling. Nat. Med. 10, 712–718. Stella, N., Schweitzer, P., and Piomelli, D. (1997). A second endogenous Justus, C.R., Dong, L., and Yang, L.V. (2013). Acidic tumor microenvironment cannabinoid that modulates long-term potentiation. Nature 388, 773–778. and pH-sensing G protein-coupled receptors. Front Physiol 4,354. Sugiura, T., Kodaka, T., Kondo, S., Nakane, S., Kondo, H., Waku, K., Ishima, Kanoh, H., Iwata, T., Ono, T., and Suzuki, T. (1986). Immunological character- Y., Watanabe, K., and Yamamoto, I. (1997). Is the cannabinoid CB1 receptor ization of sn-1,2-diacylglycerol and sn-2-monoacylglycerol kinase from pig a 2-arachidonoylglycerol receptor? Structural requirements for triggering a brain. J. Biol. Chem. 261, 5597–5602. Ca2+ transient in NG108-15 cells. J. Biochem. 122, 890–895. Kihara, Y., Maceyka, M., Spiegel, S., and Chun, J. (2014). Lysophospholipid Tager, A.M., LaCamera, P., Shea, B.S., Campanella, G.S., Selman, M., Zhao, receptor nomenclature review: IUPHAR Review 8. Br. J. Pharmacol. 171, Z., Polosukhin, V., Wain, J., Karimi-Shah, B.A., Kim, N.D., et al. (2008). The ly- 3575–3594. sophosphatidic acid receptor LPA1 links pulmonary fibrosis to lung injury by Kim, K.S., Sengupta, S., Berk, M., Kwak, Y.G., Escobar, P.F., Belinson, J., mediating fibroblast recruitment and vascular leak. Nat. Med. 14, 45–54. Mok, S.C., and Xu, Y. (2006). Hypoxia enhances lysophosphatidic acid Tanaka, M., Nakade, S., and Takaoka, Y. (2011). Compounds having lyso- responsiveness in ovarian cancer cells and lysophosphatidic acid induces phosphatidic acid receptor antagonism and uses thereof, US Patent 7875745. ovarian tumor metastasis in vivo. Cancer Res. 66, 7983–7990. Tannock, I.F., and Rotin, D. (1989). Acid pH in tumors and its potential for ther- Liu, J., Wang, L., Harvey-White, J., Osei-Hyiaman, D., Razdan, R., Gong, Q., apeutic exploitation. Cancer Res. 49, 4373–4384. Chan, A.C., Zhou, Z., Huang, B.X., Kim, H.Y., and Kunos, G. (2006). A biosyn- Tuckerman, M., Berne, B.J., and Martyna, G.J. (1992). Reversible multiple time thetic pathway for anandamide. Proc. Natl. Acad. Sci. USA 103, 13345–13350. scale molecular dynamics. J. Chem. Phys. 97, 1990–1998. Martyna, G.J., Klein, M.L., and Tuckerman, M. (1992). Nose´ –Hoover chains: Van Durme, J., Horn, F., Costagliola, S., Vriend, G., and Vassart, G. (2006). The canonical ensemble via continuous dynamics. J. Chem. Phys. 97, 2635– GRIS: glycoprotein-hormone receptor information system. Mol. Endocrinol. 2643. 20, 2247–2255. Martyna, G.J., Tobias, D.J., and Klein, M.L. (1994). Constant pressure molec- van Loenen, P.B., de Graaf, C., Verzijl, D., Leurs, R., Rognan, D., Peters, ular dynamics algorithms. J. Chem. Phys. 101, 4177–4189. S.L.M., and Alewijnse, A.E. (2011). Agonist-dependent effects of mutations

1642 Cell 161, 1633–1643, June 18, 2015 ª2015 Elsevier Inc. in the sphingosine-1-phosphate type 1 receptor. Eur. J. Pharmacol. 667, Wojtkowiak, J.W., Verduzco, D., Schramm, K.J., and Gillies, R.J. (2011). Drug 105–112. resistance and cellular adaptation to tumor acidic pH microenvironment. Mol. Wang, D.A., Lorincz, Z., Bautista, D.L., Liliom, K., Tigyi, G., and Parrill, A.L. Pharm. 8, 2032–2038. (2001). A single amino acid determines lysophospholipid speciﬁcity of the Ye, X., Hama, K., Contos, J.J., Anliker, B., Inoue, A., Skinner, M.K., Suzuki, H., S1P1 (EDG1) and LPA1 (EDG2) phospholipid growth factor receptors. Amano, T., Kennedy, G., Arai, H., et al. (2005). LPA3-mediated lysophospha- J. Biol. Chem. 276, 49213–49220. tidic acid signalling in embryo implantation and spacing. Nature 435, 104–108. Weiner, J.A., Hecht, J.H., and Chun, J. (1998). Lysophosphatidic acid receptor Yung, Y.C., Mutoh, T., Lin, M.E., Noguchi, K., Rivera, R.R., Choi, J.W., Kings- gene vzg-1/lpA1/edg-2 is expressed by mature oligodendrocytes during mye- bury, M.A., and Chun, J. (2011). Lysophosphatidic acid signaling may initiate lination in the postnatal murine brain. J. Comp. Neurol. 398, 587–598. fetal hydrocephalus. Sci. Transl. Med. 3, 99ra87.

Cell 161, 1633–1643, June 18, 2015 ª2015 Elsevier Inc. 1643